we’ve seen national ➡️ local agenda-setting in news in general, as well as in news about trans people, but the two-step information flow is a novel finding in the context of trans news! our method incorporates transfer entropy, a measure of drivenness, to get a fine-grained look at agenda-setting.

we find that agenda-setting follows a two-step flow for the sample of 2023-vintage US trans news we looked at; a lot of the time, national outlets influence one state’s local outlets which *in turn* influence what other states’ local outlets cover.

i would very much welcome your feedback on the tool, the instructions, and anything else you can think of! obviously this is very much still in development, and it will continue to evolve over the next while!!

this repo contains (minimal) code to transcribe interviews locally (i.e. on your laptop or your favorite secure server) using a model that lives on that computer. i have just added the world's worst user interface so that navigating file paths is less of a pain in the butt.

OH NO IT'S SO TRUE
(if only i had not failed band in undergrad tho)

oh hello triad closure ❤️

Of course, people don't agree on what is "safe" and "fair" and that's why moderation is hard. But you can't solve a network-level problem with individual-level tools

there are a lot of examples, even in canonical-ish works of network science or computational social science, where the stated downstream implications of data-driven analyses have so much potential for harm if taken seriously and/or circulated widely! and they often are!!

yes, that one! alas 💀

there’s a bluesky api users discord that might be a good place to lurk & learn more if you’re really invested

going profile by profile was doable for starter packs and the follower network but idk if that would work at scale for posts